suomigps: SuomiNet Global Positioning System

General Data Description

SuomiNet (named to honor meteorological satellite pioneer Verner Suomi) is a network of GPS receivers to provide real-time atmospheric precipitable water vapor measurements and other geodetic and meteorological information. ARM has built geodetic monuments for Suominet GPS receivers across the SGP CART domain.

Measurement Description

Global Positioning System (GPS) receivers and antennas provided by the Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) have been deployed at the ARM Climate Research Facility (ACRF) Tropical Western Pacific (TWP) sites at Nauru, Manus Island, and Darwin. The data from these systems are being acquired and processed by the SuomiNet project to provide 30-minute values of precipitable water vapor and other quantities listed below.

Measurements are made every half-hour; files contain one day of data.

  • Station name and network ID
  • Station latitude and logitude [degrees]
  • Station height above mean sea level (JGM2 geoid) [m]
  • Network ID

For each time period and station:

  • Duration of validity of measurement [minute]
  • Precipitable water vapor [mm]
  • Precipitable water vapor format error [mm]
  • Wet delay [mm]
  • Dry delay from model [mm]
  • Total delay [mm]
  • Pi factor (conversion between wet delay and PWV) [mm]
  • Surface atmospheric pressure [hPa]
  • Surface temperature [deg C]
  • Surface relative humidity [%]
  • Final dry delay [mm]
  • Flag to indicate whether actual or interpolated met values are used
  • Pi factor type: ‘S’ means that surface temperature coefficients were used

Temporal Coverage

Data files are available beginning 2001-06-07 for sgp and beginning 2003-02-12 for gec. Because of the redistribution of extended facilities some stations were relocated and have varying begin/end-dates (see description under Area Covered below).

The data are available at the ARM Data Center a few days after they are generated.

Area Covered

sgp30suomigpsX1.c1
The sgp datastream includes all the suominet CONUS sites.

EF Station IDs:
SGP-E1: Larned, KS (ID=SG12) Phased-out on 2010-10-15
SGP-E3: LeRoy, KS (ID=SG15) Phased-out on 2010-10-28
SGP-E5: Halstead, KS (ID=SG13) Phased-out on 2010-11-02
SGP-E6: Towanda, KS (ID=SG14) Phased-out on 2011-10-18
SGP-E7: Elk Falls, KS (ID=SG16) Phased-out on 2011-11-14
SGP-E8: Coldwater, KS (ID=SG11) Phased-out on 2010-11-10
SGP-E9: Ashton, KS (ID=SG04)
SGP-E11: Byron, OK (ID=SG10)
SGP-E12: Pawhuska, OK (ID=SG08)
SGP-E13: ARM CF Lamont, OK (ID=SG01)
SGP-E15: Ringwood, OK (ID=SG09)
SGP-E19: El Reno, OK (ID=SG20) Phased-out on 2011-05-25
SGP-E22: Cordell, OK (ID=SG19)Phased-out on 2010-12-01
SGP-E24: Cyril, OK (ID=SG18)Phased-out on 2010-11-24
SGP-E25: Seminole, OK (ID=SG17) REMOVED FROM SERVICE
SGP-E27: Earlsboro, OK (ID=) Replaced E25 at Seminole
SGP-EF31: Anthony, KS(ID=SG41).
SGP-EF32: Medford, OK(ID=SG42).
SGP-EF33: Newkirk, OK (ID=SG43). starting 2012-03-14
SGP-EF34: Maple City, KS(ID=SG44). starting 2012-03-14
SGP-EF35: Tyron, OK (ID=SG45).
SGP-EF36: Marshall, OK(ID=SG46). starting 2012-03-14
SGP-EF37: Waukomis, OK(ID=SG47).
SGP-EF38: Omega, OK(ID=SG48). starting 2012-03-14

gec30suomigpsX1.c1
The gec datastream includes all other global sites including those covering the NSA and TWP sites. See map of Global sites
ACRF site Station IDs:

 

 

 

Site ID Begin Date
SGP-E13: ARM CF Lamont, OK SG01 2003-02-12
NSA-C1: Barrow, Alaska SG27 2003-02-12
NSA-C2: Atqasuk, Alaska ATQK 2007-07-15
TWP-C1: Manus Island, Papua New Guinea SA42 2004-12-03
TWP-C3: Nauru Island, Nauru SA40 2004-11-05
TWP-C3: Darwin, Australia SA39 2004-12-01

 

Note: Even though the SGP station SG01 is included in both datastreams, the data are not identical because they are based on two different orbit solutions.

Datastream Inputs

Data files in netCDF are provided “as is” via Unidata’s LDM (Local Data Manager)

The ARM files correspond to the daily post-processed solutions, originally named “suoPWV_YEAR.DOY.HH.00.1440”. The daily files provide more accurate PWV retrievals than the hourly solutions intended for more rapid distribution of near-real-time data.

Related Links

Contacts

Vasile Tudor Garbulet

ARM Data Center

Data Source

Institution
SuomiNet Website

FAQ

Q: In the netCDF files sgpsuomigpsX1.c1.yyyymmdd.000000.cdf, each GPS station has an associated network ID, which apparently indicates which network (Suominet, FSL, DOT, …) the station belongs to. However, for the files I downloaded, I find only missing values (99.) in this field.
A: Teresa Van Hove: I have another netCDF PWV product I generate where I combine data from different hourly solutions that I run. ARM is archiving the Suomi network solution and I’m not filling in the network field for it.
However, there are a handful of sites that are operated by other agencies that I process along with the Suomi network. Some are IGS sites used to keep my reference frame current with the IGS reference frame the orbits are calculated in and others are from state agencies that use the same antenna as the Suomi network and have some kind of a met sensor (often ASOS) located near the GPS antenna.

If you want to be sure that you are only using Suomi sites, you can select sites that are SA##, SC## or SG##. I.e. CASL, HILB, IUCO are all non-SuomiNet sites and are operated by agencies in SC, NC and IL.

Data User Notes

Computing Date and Time
There are three variables in these files to compute time.
The variable for each measurement is time_offset:long_Name = “Time delta from start_time”

In addition, there are two global variables in each file:
start_date=”yyyy.DDD.00.00.00″ where DDD = Julian Day and
start_time in seconds since 1980-01-06 00:00Z (the beginning of the GPS era) which is offset 315,964,800 seconds from 1970-01-01 00:00Z (std unix time).

The time for each measurement can be computed either using time_offset and start_date or time_offset and start_time.

Missing value codes
The netCDF files were created by a conversion from an ASCII formatted intermediate file. When a station doesn’t report any values for a particular time-period all the variables are given a value of -999 for that time. In addition individual variables have different missing values present in the ASCII intermediate file that vary from variable to variable because of varying ASCII format codes:

Format of entries below is netCDF_varname(ASCII_column_header): missing_value

  • pwv(PW): -9.9
  • pwv_err(FMerr): none
  • wet_delay(Wdelay): -99.9
  • total_delay(Tdelay): -99.9
  • pifact(KFAC): -9.900
  • pres(Press): -99.9
  • temperature(Temp): -99.9
  • rh(Rhum): -99.9
  • final_dry_delay(Ddelay): -99.9
  • ASCII columns Mdelay, Mf, Kf, do not appear to have an equivalent varname in the netCDF file and appear to be always -99.9 for the ARM owned GPSes.

N.B.: The only variable where the specific missing_values matter is temperature, because all the others are always positive and all negative numbers can be treated as missing. For temperature, it would be best to replace all values -99.9 with -999 before filtering for missing_values because -999 is the designated missing_value in the netCDF attribute.

Sample Code for ACRF sites
This sample Fortran program demonstrates how to read the gec30suomigpsX1.c1 and sgp30suomigpsX1.c1 files to extract the data for the ACRF field sites.

Retrieval Process and Revision
Suominet data has been processed with the updated Bernese software. B5.0. Please click here for more information on the Suominet Bernese software update.

Acronyms

CONUS    Continental United States
GPS      Global Positioning System
PWV      Precipitable Water Vapor
SGP      Southern Great Plains

Citable References

Bevis M, S Businger, TA Herring, C Rocken, RA Anthes, and RH Ware. 1992. “GPS meteorology: Remote sensing of atmospheric water vapor using the global positioning system.” Journal of Geophysical Research 97(D14): 15,787, doi: 10.1029/92jd01517

Ware, RH, DW Fulker, SA Stein, DN Anderson, SK Avery, RD Clark, KK Droegemeier, JP Kuettner, JB Minster, and S Sorooshian. 2000. “SuomiNet: A Real-Time National GPS Network for Atmospheric Research and Education” Bulletin of the American Meteorological Society 81(4): 677-694, doi: 10.1175/1520-0477(2000)081<0677:sarngn>2.3.co;2